Professional Documents
Culture Documents
Summary
Lancet Oncol 2013; 14: 112632
Published Online
October 17, 2013
http://dx.doi.org/10.1016/
S1470-2045(13)70448-0
See Comment page 1151
Copyright Widschwendter et al.
Open Access article distributed
under the terms of CC BY
*Contributed equally
Department of Womens
Cancer, UCL EGA Institute for
Womens Health
(Prof M Widschwendter MD,
A N Rosenthal PhD,
S Philpott MSc, I Rizzuto MD,
L Fraser BSc, J Hayward MSc,
E O Fourkala PhD,
Prof A Zaikin PhD,
Prof U Menon MD,
Prof I J Jacobs MD), and
Department of Mathematics
(Prof A Zaikin PhD), University
College London, London, UK;
Barts Cancer Institute, Barts and
the London School of Medicine
and Dentistry, London, UK
(A N Rosenthal); Department of
Preventive Medicine
(M P Intermaggio MSc,
C K Edlund MSc, S J Ramus PhD,
Prof S A Gayther PhD), and
Department of Pathology
(Prof L Dubeau PhD), Keck School
of Medicine, USC/Norris
Comprehensive Cancer Center,
University of Southern
California, Los Angeles, CA, USA;
and Faculty of Medical and
Human Sciences, University of
Manchester and Manchester
Academic Health Science Centre,
Manchester, UK (Prof I J Jacobs)
Correspondence to:
Prof Martin Widschwendter,
Department of Womens Cancer,
UCL EGA Institute for Womens
Health, University College
London, London WC1E 6AU, UK
M.Widschwendter@ucl.ac.uk
1226
Background Penetrance for breast cancer, ovarian cancer, or both in carriers of BRCA1/BRCA2 mutations is
disproportionately high. Sex hormone dysregulation and altered end-organ hormone sensitivity might explain this
organ-specic penetrance. We sought to identify dierences in hormone regulation between carriers of BRCA1/2 and
women who are negative for BRCA1/2 mutations.
Methods We assessed endometrial thickness for each menstrual cycle day (as an index of hormone regulation) in
393 scans from 228 women in the UK Familial Ovarian Cancer Screening Study (UK FOCSS) known to carry either
mutation and 1573 scans from 754 women known to be negative for the mutations. To quantify dierences in
endometrial thickness we focused on days 1014 and days 2126, and calculated the area under the curve. We then
compared serum oestradiol and progesterone titres during these days of the menstrual cycle in the same groups.
Follicular and luteal oestradiol and progesterone serum titres were grouped into quartiles and odds ratios were
calculated with logistic regression.
Findings Follicular phase endometrial thickness of carriers of the mutations adjusted for age and day of the menstrual
cycle was higher (odds ratio [OR] 111, 95% CI 103120; p=00063) and luteal phase endometrial thickness lower
(090, 083098; p=0027) than for women negative for the mutations. Median luteal phase titres of progesterone
were 121% higher (p=000037) in carriers than in women negative for the mutations, and for oestradiol were 33%
higher (p=0007)ie, 59% of carriers had concentrations of serum progesterone that would have been in the top
quartile of concentrations in the control group (OR 80, 95% CI 215257; p=0008).
Interpretation Carriers of BRCA1/BRCA2 mutations are exposed to higher titres of oestradiol and progesterone
known risk-factors for breast cancer. Higher titres of oestradiol in carriers are compatible with this hormone having a
role in ovarian carcinogenesis in such women. Our ndings could not be explained by dierential contraceptive pill use.
Funding Eve Appeal, European Union, Cancer Research UK, and US National Institutes of Health.
Introduction
In all inherited cancer syndromes the germline mutation
is thought to have a so-called local eect in an organ that
is predisposed to the development of cancer, because
these mutations do not cause cancers in all organs. For
example, the increased cancer risk in carriers of the
BRCA1 and BRCA2 mutations is predominantly that of
breast cancer, ovarian cancer, or both.1 These mutations
are thought to cause cancer via a defect in DNA damage
response or in the DNA repair pathway.2 However, this
defect does not explain the organ-specic cancer
penetrance. That removal of both ovaries and Fallopian
tubes reduces not only the risk of ovarian but also breast
cancer3 implicates a systemic dysregulation of hormone
production in carriers of the mutation, which aects both
the Mllerian epithelium, as the cell of origin for ovarian
cancer,4 and breast epithelium. Evidence from preclinical
models suggests that both hormone production and
hormone sensitivity of end organs is altered in carriers of
the BRCA1 mutation. Studies in animals57 showed that
mice carrying a Brca1 mutation in the steroid-hormoneproducing granulosa cells had a longer pro-oestrus phase,
corresponding with the oestrogen-dominant follicular
Articles
Methods
Participants
After ethical approval (Eastern MREC 97/5/007),
UKFOCSS recruited from 44 UK regional centres.
Between June, 2002, and September, 2010, women older
than 35 years, at an estimated minimum 10% lifetime
risk of ovarian cancer based on family history or
predisposing germline gene mutations (appendix) were
recruited and screening data and outcomes were
collected prospectively. So far, about 25% of the study
population have undergone clinically initiated testing for
BRCA1 and BRCA2 mutations (either before or after
recruitment).
After ethical approval (Joint University College London/
University College London Hospital Ethics Committee,
ref 06/Q0505/102), we selected all premenopausal
UKFOCSS participants with no previous or subsequent
history of cancer (to avoid including women on hormonal
therapy or women with a subclinical cancer, which could
have triggered an altered hormonal environment and
which could have confounded our analyses) and no
intrauterine device.
Procedures
Ovarian cancer screening was done with transvaginal
sonography to assess ovarian morphology and serum
CA125 tumour marker measurement. All centres
scanned study participants and all scans were done by
sonographers, radiologists, or gynaecologists approved
by the UKs National Health Service (NHS), employed by
the local hospital for gynaecological scanning, and
subject to local NHS quality control. Initially, CA125 tests
were done annually, but after 2007 these tests were done
every 4 months. Samples were taken in EDTA-containing
tubes at womens primary care practices and posted to
our laboratory for aliquoting, CA125 testing, and storage
at 80C. Samples were discarded if they reached the
laboratory more than 56 h after venepuncture or were
haemolysed. Endometrial thickness and date of last
menstrual period were routinely recorded during
sonography. Information on use of oral contraceptive
www.thelancet.com/oncology Vol 14 November 2013
Statistical analysis
Articles
Results
2808 scans from 1460 UK FOCSS volunteers for whom
ET available and who do not have cancer
BRCA1 and
BRCA2
negative
BRCA1 or
BRCA2
mutation
p value*
1099 (77%)
219 (78%)
061
332 (23%)
61 (22%)
All scans
No contraceptive pill use
Contraceptive pill use
46 (81%)
191 (79%)
11 (19%)
52 (21%)
073
1053 (77%)
28 (76%)
321 (23%)
9 (24%)
089
Use of oral contraceptive pills in the same decade as endometrial thickness scan was
available for 87% of the 1966 endometrial thickness scans done. *From test.
Women whose mutation status was established with clinical genetics testing.
Women whose mutation status was established with next generation sequencing.
Articles
Discussion
Our ndings show clear dierences in cyclical
endometrial thickness in carriers of BRCA1/2 mutations
compared with wild-type controls. We have also shown
congruent changes in sex-hormone titres, which are one
of the probable explanations for this dierence. Our
cohort of well matched premenopausal carriers and
controls with combined endometrial thickness information, and serum samples with known menstrual cycle
data, was ideal for testing the hypothesis that the organspecic cancer penetrance in carriers is due to hormonal
dysregulation. To our knowledge, the UK FOCSS
population is the largest cohort for whom this
information is available (panel).
Our ndings of high luteal phase progesterone titres in
carriers of BRCA mutations, associated with decreased
endometrial thickness, are in complete concordance with
each other, and might relate to a defect in steroid-hormone
regulation in carriers that results in an end-organ eect.
We speculate that the high luteal phase oestradiol we
recorded in carriers triggers increased expression of
progesterone receptors,11 thus potentiating any possible
mutagenic eect of the higher luteal progesterone. Our
ndings support those from studies in mice carrying a
Brca1 mutation in ovarian granulosa cells.6,7 Although data
on endogenous premenopausal progesterone exposure
www.thelancet.com/oncology Vol 14 November 2013
10
B
BRCA1 and 2 negative
BRCA1 and 2 mutation
10
5
OR=111* (95% CI 103120)
p=00063
4
0
10
15
Menstrual cycle (days)
20
25
30
Articles
p value
Days 1014
of cycle
Days 2126
of cycle
AUC
AUC
40
32
31
43
BRCA2 mutation
42
32
32
41
BRCA1/2 negative
308
28
274
45
BRCA1/2 mutated
82
32
63
41
BRCA1/2 negative
308
28
274
45
<00001
<00001
p=00074
80
1000
Progesterone (nmol/L)
Oestradiol (pmol/L)
800
600
400
p=000034
60
40
20
0
Negative
BRCA1/2
mutation
Negative
BRCA1/2
mutation
Systematic review
We searched PubMed for studies on BRCA mutation and
alterations in the female premenopausal reproductive
hormone system published in English between Sept 1, 1993,
and Feb 28, 2013. We used the search terms (BRCA1 or
BRCA2) and (endocrine or hormones or endometrium
or menstrual). We established that studies in animals
showed aberrant hormonal regulation upon loss of BRCA1 in
granulosa cells and that premenopausal surgical resection of
the ovaries in women carrying the BRCA1 or BRCA2 mutation
led to a substantial reduction in the risk of breast cancer.3
Systemic endocrine eects of BRCA1 or BRCA2 mutations
have not previously been assessed in humans.
Interpretation
Our ndings suggest that BRCA1/2 germline mutations are
driving carcinogenesis only in part via altered molecular
pathways (eg, those involved in DNA repair) in the organ at
risk, and that BRCA1/2-associated changes in the endocrine
system are additional factors. These insights could act as a
major impetus for novel chemoprevention trials using
strategies that can exploit the hormonal dysregulation in
carriers of BRCA1/2 mutations. Potential agents for these
trials include selective oestrogen or progesterone receptor
modulators, and the anti-RANKL (receptor activator of NF-B
ligand) antibody denosumab (currently used for osteoporosis
treatment, but known to block the downstream carcinogenic
eects of progestogens).
200
Articles
Articles
Acknowledgments
This work was funded by the Eve Appeal and the European Unions
Seventh Framework Programme (FP7/2007-2013) under grant
agreement number 305428 (Project EpiFemCare) and done at UCLH/
UCL, which received a proportion of its funding from the Department of
Health NIHR Biomedical Research Centres (BRC) funding scheme. The
UK FOCSS was core funded by Cancer Research UK (grants C315/A2621
and C1005/A6383), the UKs Department of Health, and the Eve Appeal
with additional support from grants MRC8804/A/A7058, CA152990, and
CA086381 from the NCI Early Detection Research Network. Next
generation sequencing was done by the USC Molecular Genomic Core
and was supported in part by award number P30CA014089 from the
National Cancer Institute. The content is solely the responsibility of the
authors and does not necessarily represent the ocial views of the
National Cancer Institute or the National Institutes of Health.
LD received grant support from US National Institutes (grant numbers
R01 CA133117 and RO1 CA119078). IJJ is an NIHR Senior Investigator.
We thank David Conti for his help in the bioinformatic analysis of next
generation sequencing data.
References
1
King MC, Marks JH, Mandell JB. Breast and ovarian cancer risks
due to inherited mutations in BRCA1 and BRCA2. Science 2003;
302: 64346.
2
Venkitaraman AR. Cancer susceptibility and the functions of
BRCA1 and BRCA2. Cell 2002; 108: 17182.
3
Domchek SM, Friebel TM, Singer CF, et al. Association of
risk-reducing surgery in BRCA1 or BRCA2 mutation carriers with
cancer risk and mortality. JAMA 2010; 304: 96775.
4
Dubeau L. The cell of origin of ovarian epithelial tumours.
Lancet Oncol 2008; 9: 119197.
5
Chodankar R, Kwang S, Sangiorgi F, et al. Cell-nonautonomous
induction of ovarian and uterine serous cystadenomas in mice
lacking a functional Brca1 in ovarian granulosa cells. Curr Biol 2005;
15: 56165.
1232
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26